Sorry, I mean "separating data from presentation", not "logic from 
presentation". Obviously, even the most trivial dynamic interface will need to 
combine logic and presentation.  :-)

On Nov 22, 2010, at 8:40 AM, Greg Brown wrote:

> I think the idea of separating code and markup is different from separating 
> logic and presentation. The latter (which is common in JSP and HTML) can 
> produce unreadable and unmanageable code. However, the former can help create 
> more maintainable code. For example, if you create a button and add event 
> listeners to it in BXML, the association between the button and the listeners 
> is clearer than if you create the listeners in your initialize() method. 
> OTOH, it is also a good idea to keep event listeners as short as possible and 
> have them call into domain objects that are exposed by the application (and 
> these would generally be implemented in compiled code vs. script).
> 
> On Nov 22, 2010, at 8:26 AM, Sandro Martini wrote:
> 
>> 
>> Hi Greg,
>> you have reason, in BXML probably is less important to have this than in JSP
>> ... over years I've done too much maintenance on old-fashioned JSP.
>> 
>> But in this way we could have a simple settings to ensure that all
>> developers could use BXML only for definition of components and GUI, and be
>> sure to have have all application logic inside code. 
>> For example if a Company set a standard like this we can simplify standard
>> adherence of all Pivot-based code.
>> Of course this will be useful on complex applications, where having most of
>> the code checked at compile time will be great.
>> 
>> Do you think a ticket for this for the 2.1 will be interesting (maybe as a
>> wish, and see later if someone agree that's useful, or maybe drop it later)
>> ?
>> 
>> Comments (even from others) ? ...
>> 
>> Sandro
>> 
>> -- 
>> View this message in context: 
>> http://apache-pivot-developers.417237.n3.nabble.com/BXMLSerializer-Flag-to-turn-off-script-parsing-tp1945087p1945656.html
>> Sent from the Apache Pivot - Developers mailing list archive at Nabble.com.
> 

Reply via email to